Hello meep users,
I am a new user on meep and I am still trying on the C++
tutorial of meep in the website. The problem I faced has been posted once and
I am providing more details on the problem now.
When I am trying the "Computing the Quality
Factor of a Resonator" case with meep version 0.20.3, I get the the
following result:
time for set_epsilon = 9.5129e-05 s
creating output file "./eps-000000.00.h5"...
frequency,amplitude,quality factor
0.320941-1.00538e-07i,-0.45262-0.98774i,1.59611e+06
Field time usage:
connnecting chunks: 0.000305176 s
time stepping: 0.499064 s
communicating: 0.0733039 s
outputting fields: 0.00160599 s
Fourier transforming: 0.0106251 s
everything else: 0.0606663 s
Elapsed run time = 0.696532 s
which is a bit far from the value 3.26087e+06 posted in the website.
However
when I try with the meep 0.10 version which I get recently with the
same code (same machine also, intel with Scientific Linux), it gives
the following result:
time for set_epsilon = 9.10759e-05 s
creating output file "./eps-000000.00.h5"...
frequency,amplitude,quality factor
0.320941-4.92119e-08i,-0.35119-1.0292i,3.26081e+06
Field time usage:
connnecting chunks: 7.29561e-05 s
time stepping: 0.111602 s
communicating: 0.0455134 s
outputting fields: 0.0015049 s
Fourier transforming: 0.00917506 s
everything else: 0.0282183 s
Elapsed run time = 0.239243 s
which is much more closer to the value provided in the website.
I've also print out the parameter 'p' passing to the do_harminv function in
the last few lines of the code. The meep 0.10 is passing a set of
values to the function which is also different from the set of values
meep 0.20.3 passing to the function.
Can any one repeat my result? Will the situation be improved if I use an Unix
machine instead? Or is that a bug or is the meep 0.20.3 requiring a different
computer config compare to 0.10 version?
Woo
為了不斷提升Yahoo!
Mail，雅虎香港誠邀你參與意見調查。請前往http://surveylink.yahoo.com/wix/p5429076.aspx 發表你的意見!
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
